- Email: [email protected]
Environmental Loading Conditions for CSP Solar Fields
Available online at www.sciencedirect.com
ScienceDirect Energy Procedia 69 (2015) 1211 – 1219
International Conference on Concentrating Solar Power and Chemical Energy Systems, SolarPACES 2014
Environmental loading conditions for CSP solar fields M. Balza*, F. von Reekena a
schlaich bergermann und partner (sbp sonne GmbH), Schwabstr. 43, 70197 Stuttgart, Germany
Abstract CSP Solar Fields are designed as lightweight structures. Nevertheless, due to their size and complexity, large amounts of materials are required. Environmental conditions like wind loads are a key driver for the design of solar fields, including required strength of materials and therefore mass. To give a solid example: Structural steel work for the solar field of a 500,000 m² aperture parabolic trough plant (typical 50 MW plant in Spain with 7 h thermal storage) amounts to approx. 12,500 tons of steel. Due to changes in European design codes and the respective loading conditions, the design wind speed increased from 34 m/s to 38 m/s, which increases solar field structural steel mass to 14,450 tons, i. e. by about 16 %. Obviously, prescribed load conditions have a tremendous effect on steel mass and hence cost and environmental impact. Unfortunately, load description in many tender specifications is insufficiently accurate, what often results in confusion or even in significant errors in planning and construction, or higher risks. The paper describes various environmental loading effects, mainly wind, for parabolic trough and heliostat collector fields. © 2015 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2015 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer reviewbybythe thescientific scientific conference committee of SolarPACES 2014responsibility under responsibility Peer review conference committee of SolarPACES 2014 under of PSE AGof PSE AG. Keywords: Design Wind Speed, Environmental Loading Conditions, Techno-economical optimisation, Solar Field
* Corresponding author. Tel.: +49 711 64871 0; fax: +49 711 64871 66. E-mail address: [email protected]
1876-6102 © 2015 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer review by the scientific conference committee of SolarPACES 2014 under responsibility of PSE AG doi:10.1016/j.egypro.2015.03.211
1212
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
1. Introduction Now that more and more CSP plants are developed and built world wide the technology is becoming increasingly mature by getting exposed to changing financing conditions, other legal requirements, different utilities and clients. It seems that it is obvious that CSP projects are adopted to the local financing conditions and feed-in tariffs. Also solar resources do change from location to location and are naturally taken into consideration while optimizing the plant configurations. Having understood these facts one wonders that all other design parameters that are presumingly not directly linked to the plants financial performance are so tremendously neglected. Several power plants were built in Spain and many companies copied or developed new collector technologies. All of them are based on rather low wind speeds of 34 m/s1 related to a now long superseded Spanish wind design code. After the implementation of the Eurocodes the value was increased to 38 m/s1 in nearly whole Spain, maybe related to the known phenomenon that nowadays extreme weather conditions do occur more often due to climate change. This change in environmental conditions even lead to some moderate structural failures on some collector elements that were unfortunately also designed with insufficient safety factors. The design wind speeds in Morocco, South Africa, nearly all European Countries and many other places in the world are even more severe. However, standard collector designs suitable for 34 m/s1 with no adjustment to the much higher wind speeds are build in many places of the world. This means a collector design for South Africa with 40 m/s1 is exposed to 38 % higher wind loads than collector designs that are established for Spain with 34 m/s1, leading to dramatically reduced safety factors and highly likely failure under extreme load conditions if the same collector design is used. This simple effect is often ignored or unknown. By the very nature of lightweight structures used for solar collectors, the ratio of applied loading to self-weight is usually much larger than that of conventional building structures. Changes in the magnitude of wind and sometimes snow loading are therefore likely to have a proportionately larger impact on the size of the structural members required and the scale of deflections experienced. Consequently the selection of loading parameters (wind pressures and shape coefficients) for the design of solar collectors has to be carefully considered. Furthermore the codes are written or standardized building shapes and building behaviour, usually making the application of a single code very difficult. As a consequence more time and effort needs to be spent in defining load cases. Collector designs are to be undertaken for a key market or a specific project as on the way of making CSP affordable a “one suits all” collector design would either be conservatively too safe and therefore expensive or cheap and prone to failure. In Figure 1 a code conform wind speed is determined allowing a product design for a specific European region 26 m/s to 27 m/s (relating to approx. 38 m/s1) in green in which the structural system is perfectly utilized. Based on the laws of physics this design could be used in the yellow regions, but would need to be strengthened for the red locations – however for all non green locations a redesign would be a good idea, in order to save costs or achieve acceptable levels of safety.
1=
3 sec. gust wind speed , measured at 10 m height, return period 50 years
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
1213
Fig. 1. Simplified wind speed for 26 m/s to 27 m/s (relating to approx. 38 m/s1) in green, >27 m/s in red, < 26 m/s in yellow based on design codes for 10 min mean, 10 m height, 50 years return period
2. Adaption engineering Based on different locations in the world, various environmental conditions will call for an adoption of the already tried and tested “bankable” collector technology due to the considerable differences in environmental conditions. The list below highlights differences or room for improvement for correctly designed European collector designs. The list is a general example and not complete: x
Higher Design Wind Speeds – extreme wind conditions (i.e. South Africa, Morocco, Rajasthan, Gujarat, Florida, Greece many locations)
x
Directionality of Design Wind Speeds – extreme wind conditions (i. e. South Africa, India)
x
Lower Design Wind Speeds – extreme wind conditions (i. e. China, Egypt….)
x
Lower Operational Wind Speeds – extreme wind conditions (i. e. South Africa)
x
Extreme low ambient temperatures (i. e. China / Inner Mongolia)
x
Seismic actions (i.e. California, Chile, Greece, Turkey)
x
Corrosiveness of local environment (Shoreline locations, industrial areas and ISCC plants)
1214
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
x
Geotechnical Requirements for stiffness, strength and type of foundations (all locations)
x
Snow or Ice loading (European locations and others)
All conditions which are listed above should be considered in order to subsequently adapt the collector technology to the specific site characteristics, thus ensure savings in some cases up to thousands of tons of steel or increase levels of safety. In order to enable an effective adaption engineering, the redesign time of usually less than six weeks and a flexible designed collector technology (changes in member sizes, wall thicknesses of sections etc.) is indispensable. From a techno-economical side of view some collector technologies may not be suitable for adaption to very high wind speeds as the increase of wall thickness (i.e. stamped parts) might not be sufficient. Also specific parts of collectors are designed for gentle corrosive environment and may require complete material changes. 3. Definition of environmental loading conditions - Wind Unfortunately, the basis of concentrator design codes in tender documents for many projects worldwide is not well defined. Moreover, design wind speeds specified in tender specifications often lack scientific background or references to corresponding building codes. This leads to significant uncertainties and misunderstandings and finally to collector structures that are either too strong – and hence expensive – or, in the worst case, too weak increasing the risk of failure. Therefore, a precise and detailed load definition is fundamental to obtain a fair allocation and a safe product, i. e. a safe and fully operational solar field. Building codes follow a certain safety concept. Hence, the actions are defined with a certain probability of occurrence, such as wind speed with an annual exceedance probability of 2 % (corresponding to a mean return period of 50 years) [1]. The design code then defines safety factors for the strength of support structures. In order to follow a consistent safety concept, the probability of occurrence of the actions should correspond to the safety factors of the design code. In addition to the site specific local codes a statistical extreme wind analysis should be undertaken in order to prove the accuracy of codes, reduce the design wind speed stipulated by the code and/or reduce risk in load assumptions. This analysis is not always leading to lower design wind speeds compared to the codes but the additional security might also serve to decrease levels of safety on a scientific basis. Providing sufficient data, a directional study of wind loads might be undertaken utilizing the knowledge that the extreme wind (not prevailing wind) does also come from specific directions only.
Fig. 2: Anemometer with wind vane and long term weather station at airport location
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
1215
“Wind speeds” are often provided in owners documents for the plant to be developed, sometimes even obtained by weather stations on site, but these are unfortunately completely worthless for the design of structures due to: x
Missing measurement duration (extreme wind analysis >20 years)
x
Missing definition of measurement basis (World Meterological Organization standards, anemometer height, directions etc.)
x
Missing scientific evaluation basis (gust wind speed i. e. 3 sec, mean wind speed 10 min mean, reoccurrence period)
As an example for a mayor CSP plant in the MENA region a design wind speed was stipulated stating “35 m/s (3sec)” in the tender documentation with no further description except that the value should be checked. A statistical analysis of the design wind speed was undertaken by one of the bidders which lead to a design wind pressure that was 25% higher. Taking the correct values the solar field rendered too expensive and most likely the “cheap” but insufficiently designed solar field will be chosen – resulting in considerable risk of failure. 4. Peak Wind Velocity Pressure Having defined the basic wind speed the codes [1] do allow to define the design relevant peak wind velocity pressure for static or even dynamic design. The peak wind velocity pressure is taken into account: x
Directional factors
x
Terrain roughness
x
Height of structure
x
Turbulence intensity
x
Air density (height)
Multiplying the peak wind velocity pressures with shape coefficients called cp-factors the structural system can be designed analyzed and optimized. The cp-factors are based on general wind tunnel studies [2] [3] for simple collector designs or specific and extensive wind tunnel studies for advanced collector technologies [4] even reducing cp-factor by wind ventilation gaps or specific shapes. The published wind tunnel data may not be suitable [2] [3] in collector shapes, support structures and solar field layouts that vary of the investigated. 5. Relevant wind speeds for the design of CSP plants In order to clarify the requirements of the different wind data information for a plant, the following description is provided:
1216
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
Table 1. Wind data required for CSP field design.
Common Name
Description
Relevant for
Prevailing Winds and Directions
Wind rose, probability tables based on site measurements of around one year or more
Overall design of plant (Receiver CRS, Cooling system)
Mean Wind Speeds
Probability tables based on site measurements of around one year or more
Overall design of plant (Receiver CRS, Cooling system)
Measurement Data (Site)
Real data measurement recording (15 min, hourly, preferably with directions and description of type of data recorded) based on site measurements of around one year or more – raw data
Analysis of statistically operational wind speeds and alarm wind speeds for design of solar field
Preferably in accordance to Eurocode 1: Actions on structures Part 1-4: General actions - Wind actions called Fundamental Basic Wind Velocity: The 10 min mean wind velocity with an annual risk being exceeded of 0.02, irrespectively of wind direction, at a height 10m above flat open country terrain and accounting for altitude effects (if required)
Structural wind load determination by design engineer solar field and therefore structural design
Design Wind Speed
Receiver losses CRS plants
Or ASCE code equivalent Long Term Measurement Data (Meteorological Measurements Station, Airports etc. to WMO standards )
Real data measurement recording (15min, hourly, gust recordings with directions and description of type of data recorded) based on weather station measurements of more than 20 complete years – raw data
Statistical Analysis of Design Wind Speed by Wind Engineer using Gumbel or Fisher Tippet extreme value distribution, preferably directional all leading to an improved Design Wind Speed
1217
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
6. Relevant Wind Speeds for structural design of solar collector In stark contrast to all building codes CSP collectors do utilize their ability to move into stow position to decrease wind loading. The relevant wind loading for such operational conditions underlies a techno-economical approach maximizing solar operating hours while at the same time minimizing cost of structure, drive costs and foundation expenses. Here the local wind profile also plays a major role. As an example most European climates have higher probability of lower wind speeds. Therefore operational wind speeds are generally higher, causing higher cost for drives, drive energy consumption and requiring stiffer support systems to achieve good intercept factors and the cheapest optimized LCOE. Most desert locations show higher extreme wind conditions, but at the same time lower operational wind speeds leading to cheaper drive systems but higher strength requirements on all structural parts in stow position. Table 2. Description and applicability of three main wind speeds for CSP plant design and operation.
Design Goal
Operational wind speed
“Go to stow” wind speed
Structural Design Wind Speed
Economic Operation
Ultimate Limit State Design for Structure and Drives in all (relevant) positions
Ultimate Limit State Design for Structure and Drives for stow position only
5-8
9-17
25-37
Approx. 5 %
Approx. 25 %
100 %
Services Limit State Design Magnitude of wind speed (m/s 10 min mean, 10 m height, 50 years return period for design wind speed) – all site and technology dependent Magnitude of design wind pressure (percentage ) – all site and technology dependent
1218
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
7. Statistical Directional Design Wind Analysis Only very sophisticated codes do cover the wind load determination utilizing the fact that the higher wind speeds often have typical directions2 especially in desert locations. For parabolic trough plant solar fields that are orientated north/south significant wind load reductions might be utilized taking into account directional methods if the extreme winds coincide with northerly/southerly directions.
Fig. 3: Directional peak wind velocity pressure [kN/m²] in accordance to codes and directional methods[2]
8. Risk management Facing conservative building codes that usually ensure safety for the build environment we live in, the slight reduction of levels of safety is in view of the writer for non occupational solar fields acceptable. Some design codes (ASCE, EC) do allow for those kinds of slight reductions. For insufficiently strong collector designs the following situation is described. Since it is not expected that the design load (typically corresponding to a recurring period of 50 years or more) occurs within the warranty period, the majority of the risk lies with the owner or his insurance company. With decreasing safety levels, increasing Operational &Maintenance costs, costs for repair of damage or even complete loss will accrue, which often have to be covered by the insurance of the owner. Hence, the insurance costs will rise with the next power plant – not only for the particular power plant, but CSP plants in general. In order to assess or to define the risk level accurately for tender, an Owners Engineer (OE) experienced in CSP and environmental loading conditions should be involved and aware of this key issue. It is also compulsory that the OE is required to check the planning at later phases in order to ensure the stipulated levels of safety. This is especially required in countries where no external control facility is established, like for example the Chief Building Officer (CBO) in the US. 2
= note extreme wind directions may not bear any relations with prevailing wind directions
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
1219
9. Conclusion It needs to be stated that the current situation of insufficient collector technologies being used at non suitable locations will consequently lead to failures of parts of solar fields. Some smaller failures of wind induced falling over parabolic trough collectors and partial failures of parabolic trough cantilevers arms due to stability problems already happened. While cost reductions are essential to make CSP competitive this will result much more out of clever engineering and techno-economical evaluation rather than saving the adaption engineering fee for a new site with completely different boundary conditions. In order to define an acceptable limit of risk at the smallest expense and to define state of the art design methods for CSP solar fields, a common design code is required. With a short sighted view the current situation allows the possibility to win a project due to cheap but unfortunately not fit for use solar fields. This consequently leads to high risks for owners and insurers. For each project the client or its owners engineer should define and enforce acceptable limits of safety providing scientific basis and background. While in many countries the local codes might not even cover the site locations nor the accuracy of loading required for such large structures (i.e. cost), additional efforts (i.e. statistical environmental load studies wind, maybe snow) are compulsory. It is advised that the owners engineer shall have in depth knowledge of collector design not only covering thermal output of solar collectors or optical requirements of heliostat, but also profound knowledge of structural design – taking into account that the impact on costs and risk of a solar field is significant. References [1] Eurocode 1, 2010. Actions on structures - Part 1-4: General actions - Wind actions, German version EN 1991-1-4: 2005 + AC: 2010 [2] Peterka, J. A. & al, 2008. Wind Tunnel Test on Parabolic Trough Solar Collectors: NREL. [3] Peterka, J. & al, 1992. Wind Load Design Methos for ground based heliostats and Parabolic Dish Collectors: Sandia. [4] Balz, M., Keck, T., Schiel, W. & Weinrebe, G., 2012. Sun, Stuttgart: schlaich bergermann und partner.
ScienceDirect Energy Procedia 69 (2015) 1211 – 1219
International Conference on Concentrating Solar Power and Chemical Energy Systems, SolarPACES 2014
Environmental loading conditions for CSP solar fields M. Balza*, F. von Reekena a
schlaich bergermann und partner (sbp sonne GmbH), Schwabstr. 43, 70197 Stuttgart, Germany
Abstract CSP Solar Fields are designed as lightweight structures. Nevertheless, due to their size and complexity, large amounts of materials are required. Environmental conditions like wind loads are a key driver for the design of solar fields, including required strength of materials and therefore mass. To give a solid example: Structural steel work for the solar field of a 500,000 m² aperture parabolic trough plant (typical 50 MW plant in Spain with 7 h thermal storage) amounts to approx. 12,500 tons of steel. Due to changes in European design codes and the respective loading conditions, the design wind speed increased from 34 m/s to 38 m/s, which increases solar field structural steel mass to 14,450 tons, i. e. by about 16 %. Obviously, prescribed load conditions have a tremendous effect on steel mass and hence cost and environmental impact. Unfortunately, load description in many tender specifications is insufficiently accurate, what often results in confusion or even in significant errors in planning and construction, or higher risks. The paper describes various environmental loading effects, mainly wind, for parabolic trough and heliostat collector fields. © 2015 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2015 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer reviewbybythe thescientific scientific conference committee of SolarPACES 2014responsibility under responsibility Peer review conference committee of SolarPACES 2014 under of PSE AGof PSE AG. Keywords: Design Wind Speed, Environmental Loading Conditions, Techno-economical optimisation, Solar Field
* Corresponding author. Tel.: +49 711 64871 0; fax: +49 711 64871 66. E-mail address: [email protected]
1876-6102 © 2015 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer review by the scientific conference committee of SolarPACES 2014 under responsibility of PSE AG doi:10.1016/j.egypro.2015.03.211
1212
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
1. Introduction Now that more and more CSP plants are developed and built world wide the technology is becoming increasingly mature by getting exposed to changing financing conditions, other legal requirements, different utilities and clients. It seems that it is obvious that CSP projects are adopted to the local financing conditions and feed-in tariffs. Also solar resources do change from location to location and are naturally taken into consideration while optimizing the plant configurations. Having understood these facts one wonders that all other design parameters that are presumingly not directly linked to the plants financial performance are so tremendously neglected. Several power plants were built in Spain and many companies copied or developed new collector technologies. All of them are based on rather low wind speeds of 34 m/s1 related to a now long superseded Spanish wind design code. After the implementation of the Eurocodes the value was increased to 38 m/s1 in nearly whole Spain, maybe related to the known phenomenon that nowadays extreme weather conditions do occur more often due to climate change. This change in environmental conditions even lead to some moderate structural failures on some collector elements that were unfortunately also designed with insufficient safety factors. The design wind speeds in Morocco, South Africa, nearly all European Countries and many other places in the world are even more severe. However, standard collector designs suitable for 34 m/s1 with no adjustment to the much higher wind speeds are build in many places of the world. This means a collector design for South Africa with 40 m/s1 is exposed to 38 % higher wind loads than collector designs that are established for Spain with 34 m/s1, leading to dramatically reduced safety factors and highly likely failure under extreme load conditions if the same collector design is used. This simple effect is often ignored or unknown. By the very nature of lightweight structures used for solar collectors, the ratio of applied loading to self-weight is usually much larger than that of conventional building structures. Changes in the magnitude of wind and sometimes snow loading are therefore likely to have a proportionately larger impact on the size of the structural members required and the scale of deflections experienced. Consequently the selection of loading parameters (wind pressures and shape coefficients) for the design of solar collectors has to be carefully considered. Furthermore the codes are written or standardized building shapes and building behaviour, usually making the application of a single code very difficult. As a consequence more time and effort needs to be spent in defining load cases. Collector designs are to be undertaken for a key market or a specific project as on the way of making CSP affordable a “one suits all” collector design would either be conservatively too safe and therefore expensive or cheap and prone to failure. In Figure 1 a code conform wind speed is determined allowing a product design for a specific European region 26 m/s to 27 m/s (relating to approx. 38 m/s1) in green in which the structural system is perfectly utilized. Based on the laws of physics this design could be used in the yellow regions, but would need to be strengthened for the red locations – however for all non green locations a redesign would be a good idea, in order to save costs or achieve acceptable levels of safety.
1=
3 sec. gust wind speed , measured at 10 m height, return period 50 years
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
1213
Fig. 1. Simplified wind speed for 26 m/s to 27 m/s (relating to approx. 38 m/s1) in green, >27 m/s in red, < 26 m/s in yellow based on design codes for 10 min mean, 10 m height, 50 years return period
2. Adaption engineering Based on different locations in the world, various environmental conditions will call for an adoption of the already tried and tested “bankable” collector technology due to the considerable differences in environmental conditions. The list below highlights differences or room for improvement for correctly designed European collector designs. The list is a general example and not complete: x
Higher Design Wind Speeds – extreme wind conditions (i.e. South Africa, Morocco, Rajasthan, Gujarat, Florida, Greece many locations)
x
Directionality of Design Wind Speeds – extreme wind conditions (i. e. South Africa, India)
x
Lower Design Wind Speeds – extreme wind conditions (i. e. China, Egypt….)
x
Lower Operational Wind Speeds – extreme wind conditions (i. e. South Africa)
x
Extreme low ambient temperatures (i. e. China / Inner Mongolia)
x
Seismic actions (i.e. California, Chile, Greece, Turkey)
x
Corrosiveness of local environment (Shoreline locations, industrial areas and ISCC plants)
1214
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
x
Geotechnical Requirements for stiffness, strength and type of foundations (all locations)
x
Snow or Ice loading (European locations and others)
All conditions which are listed above should be considered in order to subsequently adapt the collector technology to the specific site characteristics, thus ensure savings in some cases up to thousands of tons of steel or increase levels of safety. In order to enable an effective adaption engineering, the redesign time of usually less than six weeks and a flexible designed collector technology (changes in member sizes, wall thicknesses of sections etc.) is indispensable. From a techno-economical side of view some collector technologies may not be suitable for adaption to very high wind speeds as the increase of wall thickness (i.e. stamped parts) might not be sufficient. Also specific parts of collectors are designed for gentle corrosive environment and may require complete material changes. 3. Definition of environmental loading conditions - Wind Unfortunately, the basis of concentrator design codes in tender documents for many projects worldwide is not well defined. Moreover, design wind speeds specified in tender specifications often lack scientific background or references to corresponding building codes. This leads to significant uncertainties and misunderstandings and finally to collector structures that are either too strong – and hence expensive – or, in the worst case, too weak increasing the risk of failure. Therefore, a precise and detailed load definition is fundamental to obtain a fair allocation and a safe product, i. e. a safe and fully operational solar field. Building codes follow a certain safety concept. Hence, the actions are defined with a certain probability of occurrence, such as wind speed with an annual exceedance probability of 2 % (corresponding to a mean return period of 50 years) [1]. The design code then defines safety factors for the strength of support structures. In order to follow a consistent safety concept, the probability of occurrence of the actions should correspond to the safety factors of the design code. In addition to the site specific local codes a statistical extreme wind analysis should be undertaken in order to prove the accuracy of codes, reduce the design wind speed stipulated by the code and/or reduce risk in load assumptions. This analysis is not always leading to lower design wind speeds compared to the codes but the additional security might also serve to decrease levels of safety on a scientific basis. Providing sufficient data, a directional study of wind loads might be undertaken utilizing the knowledge that the extreme wind (not prevailing wind) does also come from specific directions only.
Fig. 2: Anemometer with wind vane and long term weather station at airport location
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
1215
“Wind speeds” are often provided in owners documents for the plant to be developed, sometimes even obtained by weather stations on site, but these are unfortunately completely worthless for the design of structures due to: x
Missing measurement duration (extreme wind analysis >20 years)
x
Missing definition of measurement basis (World Meterological Organization standards, anemometer height, directions etc.)
x
Missing scientific evaluation basis (gust wind speed i. e. 3 sec, mean wind speed 10 min mean, reoccurrence period)
As an example for a mayor CSP plant in the MENA region a design wind speed was stipulated stating “35 m/s (3sec)” in the tender documentation with no further description except that the value should be checked. A statistical analysis of the design wind speed was undertaken by one of the bidders which lead to a design wind pressure that was 25% higher. Taking the correct values the solar field rendered too expensive and most likely the “cheap” but insufficiently designed solar field will be chosen – resulting in considerable risk of failure. 4. Peak Wind Velocity Pressure Having defined the basic wind speed the codes [1] do allow to define the design relevant peak wind velocity pressure for static or even dynamic design. The peak wind velocity pressure is taken into account: x
Directional factors
x
Terrain roughness
x
Height of structure
x
Turbulence intensity
x
Air density (height)
Multiplying the peak wind velocity pressures with shape coefficients called cp-factors the structural system can be designed analyzed and optimized. The cp-factors are based on general wind tunnel studies [2] [3] for simple collector designs or specific and extensive wind tunnel studies for advanced collector technologies [4] even reducing cp-factor by wind ventilation gaps or specific shapes. The published wind tunnel data may not be suitable [2] [3] in collector shapes, support structures and solar field layouts that vary of the investigated. 5. Relevant wind speeds for the design of CSP plants In order to clarify the requirements of the different wind data information for a plant, the following description is provided:
1216
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
Table 1. Wind data required for CSP field design.
Common Name
Description
Relevant for
Prevailing Winds and Directions
Wind rose, probability tables based on site measurements of around one year or more
Overall design of plant (Receiver CRS, Cooling system)
Mean Wind Speeds
Probability tables based on site measurements of around one year or more
Overall design of plant (Receiver CRS, Cooling system)
Measurement Data (Site)
Real data measurement recording (15 min, hourly, preferably with directions and description of type of data recorded) based on site measurements of around one year or more – raw data
Analysis of statistically operational wind speeds and alarm wind speeds for design of solar field
Preferably in accordance to Eurocode 1: Actions on structures Part 1-4: General actions - Wind actions called Fundamental Basic Wind Velocity: The 10 min mean wind velocity with an annual risk being exceeded of 0.02, irrespectively of wind direction, at a height 10m above flat open country terrain and accounting for altitude effects (if required)
Structural wind load determination by design engineer solar field and therefore structural design
Design Wind Speed
Receiver losses CRS plants
Or ASCE code equivalent Long Term Measurement Data (Meteorological Measurements Station, Airports etc. to WMO standards )
Real data measurement recording (15min, hourly, gust recordings with directions and description of type of data recorded) based on weather station measurements of more than 20 complete years – raw data
Statistical Analysis of Design Wind Speed by Wind Engineer using Gumbel or Fisher Tippet extreme value distribution, preferably directional all leading to an improved Design Wind Speed
1217
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
6. Relevant Wind Speeds for structural design of solar collector In stark contrast to all building codes CSP collectors do utilize their ability to move into stow position to decrease wind loading. The relevant wind loading for such operational conditions underlies a techno-economical approach maximizing solar operating hours while at the same time minimizing cost of structure, drive costs and foundation expenses. Here the local wind profile also plays a major role. As an example most European climates have higher probability of lower wind speeds. Therefore operational wind speeds are generally higher, causing higher cost for drives, drive energy consumption and requiring stiffer support systems to achieve good intercept factors and the cheapest optimized LCOE. Most desert locations show higher extreme wind conditions, but at the same time lower operational wind speeds leading to cheaper drive systems but higher strength requirements on all structural parts in stow position. Table 2. Description and applicability of three main wind speeds for CSP plant design and operation.
Design Goal
Operational wind speed
“Go to stow” wind speed
Structural Design Wind Speed
Economic Operation
Ultimate Limit State Design for Structure and Drives in all (relevant) positions
Ultimate Limit State Design for Structure and Drives for stow position only
5-8
9-17
25-37
Approx. 5 %
Approx. 25 %
100 %
Services Limit State Design Magnitude of wind speed (m/s 10 min mean, 10 m height, 50 years return period for design wind speed) – all site and technology dependent Magnitude of design wind pressure (percentage ) – all site and technology dependent
1218
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
7. Statistical Directional Design Wind Analysis Only very sophisticated codes do cover the wind load determination utilizing the fact that the higher wind speeds often have typical directions2 especially in desert locations. For parabolic trough plant solar fields that are orientated north/south significant wind load reductions might be utilized taking into account directional methods if the extreme winds coincide with northerly/southerly directions.
Fig. 3: Directional peak wind velocity pressure [kN/m²] in accordance to codes and directional methods[2]
8. Risk management Facing conservative building codes that usually ensure safety for the build environment we live in, the slight reduction of levels of safety is in view of the writer for non occupational solar fields acceptable. Some design codes (ASCE, EC) do allow for those kinds of slight reductions. For insufficiently strong collector designs the following situation is described. Since it is not expected that the design load (typically corresponding to a recurring period of 50 years or more) occurs within the warranty period, the majority of the risk lies with the owner or his insurance company. With decreasing safety levels, increasing Operational &Maintenance costs, costs for repair of damage or even complete loss will accrue, which often have to be covered by the insurance of the owner. Hence, the insurance costs will rise with the next power plant – not only for the particular power plant, but CSP plants in general. In order to assess or to define the risk level accurately for tender, an Owners Engineer (OE) experienced in CSP and environmental loading conditions should be involved and aware of this key issue. It is also compulsory that the OE is required to check the planning at later phases in order to ensure the stipulated levels of safety. This is especially required in countries where no external control facility is established, like for example the Chief Building Officer (CBO) in the US. 2
= note extreme wind directions may not bear any relations with prevailing wind directions
M. Balz and F. von Reeken / Energy Procedia 69 (2015) 1211 – 1219
1219
9. Conclusion It needs to be stated that the current situation of insufficient collector technologies being used at non suitable locations will consequently lead to failures of parts of solar fields. Some smaller failures of wind induced falling over parabolic trough collectors and partial failures of parabolic trough cantilevers arms due to stability problems already happened. While cost reductions are essential to make CSP competitive this will result much more out of clever engineering and techno-economical evaluation rather than saving the adaption engineering fee for a new site with completely different boundary conditions. In order to define an acceptable limit of risk at the smallest expense and to define state of the art design methods for CSP solar fields, a common design code is required. With a short sighted view the current situation allows the possibility to win a project due to cheap but unfortunately not fit for use solar fields. This consequently leads to high risks for owners and insurers. For each project the client or its owners engineer should define and enforce acceptable limits of safety providing scientific basis and background. While in many countries the local codes might not even cover the site locations nor the accuracy of loading required for such large structures (i.e. cost), additional efforts (i.e. statistical environmental load studies wind, maybe snow) are compulsory. It is advised that the owners engineer shall have in depth knowledge of collector design not only covering thermal output of solar collectors or optical requirements of heliostat, but also profound knowledge of structural design – taking into account that the impact on costs and risk of a solar field is significant. References [1] Eurocode 1, 2010. Actions on structures - Part 1-4: General actions - Wind actions, German version EN 1991-1-4: 2005 + AC: 2010 [2] Peterka, J. A. & al, 2008. Wind Tunnel Test on Parabolic Trough Solar Collectors: NREL. [3] Peterka, J. & al, 1992. Wind Load Design Methos for ground based heliostats and Parabolic Dish Collectors: Sandia. [4] Balz, M., Keck, T., Schiel, W. & Weinrebe, G., 2012. Sun, Stuttgart: schlaich bergermann und partner.